Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Research Methods > L1+2- General Underlying Principles > Flashcards

L1+2- General Underlying Principles Flashcards

1
Q

What are the 5 crucial things in the structure of all psychological research?

A
  • clear RATIONALE
  • RESEARCH QUESTION
  • RESEARCH DESIGN
  • INFERENTIAL ANALYSIS
  • EVALUATION and INTERGRATION
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are good features of a research question?

A
  • contains an “?” at the end
  • lists all constructs
  • indicates relavent populations
  • specifies relationships among constructs
  • clearly be derived from rationale for research
  • imply research design for collecting data
  • indicate kind of analysis
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What does an association imply?

A

correlation or contingency table

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What does a prediction imply?

A

Regression!

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What does group difference imply?

A

t-test, ANOVA or linear contrast

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is a construct?

A

An unobservable attribute of people that we use in both theories and research to explain human behaviour, cognition and affect. Can’t directly measure this.

Use a construct measure

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is a construct score?

A

A score on a construct measure is typically a numerical value, assigned to an individual by the method of measurement.

Eg: summed total on a questionnaire, milliseconds for a reaction time, or freq count.

these scaling units are often ARBITRARY, with NO TRUE ZERO

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is a transformed score?

A

A score that is obtained by applying any kind of mathematical formula to the raw score.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is a deviation score?

A

A common transformed score, that is obtained by subtracting the mean value of all scores to each person’s raw score.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is a standardised score?

A

a raw score transformed into a new value that has:

  1. a predefined mean
  2. a predefined scale/metric based on number of raw units equated to each SD.

eg. Z scores

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What are z-scores?

A

a special standardised score.
mean = 0
SD = 1

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What are examples of a summary feature?

A

Mean
Correlation
Diff. between means of two groups
Regression coefficient

We use these to make an inference, not individual scores.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is a parameter?

A

= POPULATION parameter

a summary characteristic of a population that is almost always unknown in practice.

exception is IQ.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is a statistic?

A

= SAMPLE statistic

a numerical value for a summary characteristic calculated on scores from one sample.
it is usually used as an estimate for a corresponding unknown pop parameter.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is a sampling distribution?

A

A distribution formed using sample statistics as random variables, using random samples.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Features of a sampling dist?

A

SHAPE:
Depends on the statistic being used, size of sample being used and underlying dist of the population

MEAN:
If unbiased statistic, it should equal the corresponding pop parameter.

SD:
This is called the STANDARD ERROR.

as sample gets larger, it becomes more normal and should approximate the theoretical probability distribution.

17
Q

What is an observed test statistic?

A

A type of standardising transformation that calculates SAMPLE STATISTIC to OBS. TEST STATISTIC.

found by subtracting observed statstic from null assumed pop parameter, and divided by standard error

doing this allows us to compare sample stat to theoretical distribution, to make inferences about unknown pop parameters.

18
Q

What is the deviation score used to calculate?

A

It is used in the numerator for observed test score, and z scores.

19
Q

What is a theoretical probability distribution?

A

It is a distribution of values for a sample statistic, and indicates the probability of observing a particular value for a sample state for specific parameter values.

this is the same as a STANDARDISED SAMPLING DIST.

thus, we do not need to construct a sampling dist.

20
Q

How do we standardise sample statistics?

A
  1. Subtract the known population parameter value to centre the sampling dist around zero
  2. Standardise the sampling dist by dividing the deviation form statstic value by standard error

THIS CREATES A THEORETICAL PROB DISTRIBUTION.

21
Q

What’s the difference between standardised sampling dist compared to non-standardised sampling dist, in shape?

A

Standardised form is fatter, and less tall – > more variation.

mean = 0
SD = 1

they are Z STATS!!

22
Q

What are the key properties of estimators?

A

Bias
Consistency
Efficiency

23
Q

What is referred to as the consistency of an estimator?

A

How well an estimator continuously gets closer to the true population parameter value as the sample size increases.

this is the saving grace of biased estimators!!

24
Q

What is referred to as the efficiency of an estimator?

A

The degree of variation of an estimator from the population para value.
smallest variation/standard error = more efficient.

25
Q

What is referred to an unbiased estimator?

A

If the mean of the sampling dist of a sample stat equals the pop para value.

26
Q

What are the 6 steps to performing a NHST?

A
  1. Specify a null hypo, and an alternative hypo
  2. decide an alpha value
  3. obtain observed construct scores, and calculate sample stat and standard error
  4. calculate observed test statistic for the null hypo
  5. find p value… etc
27
Q

What are the two diff ways you can approach a NHST?

A
  1. compare the p value obtained from the hypothesis test to the alpha value
  2. compare the observed test statistic to a critical test statistic value.
28
Q

What is the obtained p value?

A

This is the probability of getting the observed test statistic, given that the null hypothesised value is true.

Conditional probability.

Pr( Tobs | Null true)

.. it doesn’t really tell us anything about the truth of the null hypo.

29
Q

What is the underlying reasoning behind NHST.

A

We can reject the null hypothesis when the sample statistic is located further and further away from the assumed population value (mean in sampling dist), as it is reasonable to infer that the sample statistic value is very rare, OR the assumed pop is incorrect (but this is weak because need real proof).

30
Q

What is the meaning of the alpha criterion?

A

It defines the cut point at which the obtained p value for an observed test stat is viewed as sufficiently unlikely to support assuming that the null hypo is true,

it is defined as Pr ( rejecting null | null is true). It is a conditional prob, and defines a region of rejection.

31
Q

What is the generic formula for calculating a confidence interval?

A

lower bound = observed sample stat - (ALPHA VALUE x Standard error)

upper bound: observed sample stat + (ALPHA VALUE x Standard error)

32
Q

What is it called when you obtained p value against null hypo value?

A

P value function plot

you can make this by changing the hypothesised null value, and running numbers NHSTs.

Research Methods (11 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions